Setup

library(jsonlite)
library(dplyr)
library(janitor)
library(lubridate) # for ymd_hms()
library(leaflet)
library(htmltools) #for htmlEscape()
library(glue)
library(tidyr)
library(purrr)
library(sf)
library(BelgiumMaps.StatBel)

# fetch all sensor data from Luftdaten API
sensor_data <- fromJSON('http://api.luftdaten.info/static/v1/data.json', flatten = TRUE) %>%
  as_tibble() %>%
  clean_names()

# fetch data for single sensor
# sensor <- fromJSON('http://api.luftdaten.info/v1/sensor/8993/', flatten = TRUE) %>%
#   as_tibble() %>%
#   clean_names()

# fix some variable types
sensor_data <- sensor_data %>%
  mutate(
    timestamp = ymd_hms(timestamp), 
    location_latitude = as.numeric(location_latitude),
    location_longitude = as.numeric(location_longitude),
    location_altitude = as.numeric(location_altitude),
    sensor_pin = as.integer(sensor_pin))

Map of worldwide Luftdaten sensors

# helper function to past measurment vars to single string
paste_measurement_str <- function(measurements) {
  measurement_str <- measurements %>%
    mutate(measurement_str = paste0(value, " (", value_type, ")")) %>%
    pull(measurement_str) %>%
    paste(collapse = ', ')  
  
  return(measurement_str)
}

sensor_data_unique <- sensor_data %>% 
  distinct(sensor_id, .keep_all = TRUE) %>%
  mutate(measurement_str = map(sensordatavalues, paste_measurement_str)) %>%
  mutate(popup_html = glue('<strong>Sensor ID</strong>: {sensor_id}<br/>
                           <strong>Location ID</strong>: {location_id}<br/>
                           <strong>Timestamp</strong>: {timestamp}<br/>
                           <strong>Measurements</strong>: {measurement_str}<br/>'))

# set custom marker-icon from Font Awesome library
icons <- awesomeIcons(
  icon = 'fa-thermometer-half',
  iconColor = 'black',
  library = 'fa'
)

m_sensors <- leaflet(sensor_data_unique, width = '100%') %>%
  addTiles() %>%
  addAwesomeMarkers(
    lng = ~location_longitude, lat = ~location_latitude,
    popup = ~popup_html,
    icon = icons,
    clusterOptions = markerClusterOptions())

m_sensors

Voronoi map of fine dust readings in Brussels

data("BE_ADMIN_REGION")
region <- st_as_sf(BE_ADMIN_REGION)
bxl <- region %>% filter(CD_RGN_REFNIS == "4000")

https://stackoverflow.com/a/55236074/125085

measurements_data <- sensor_data %>%
  unnest() %>%
  distinct(location_id, sensor_id, value_type, .keep_all = TRUE) %>%
  mutate(value = as.numeric(value))

measurements_data <- st_as_sf(
  measurements_data, coords = c("location_longitude", "location_latitude"), 
  crs = 4326, agr = "constant")

bxl_measurements <- measurements_data %>%
  filter(lengths(st_within(measurements_data, bxl)) == 1) %>%
  filter(value_type == 'P1') # se

bxl_voronoi <- bxl_measurements  %>% 
  st_union() %>%
  st_voronoi() %>%
  st_collection_extract()

leaflet(bxl_voronoi, width = '100%') %>%
  addTiles() %>%
  addPolygons() %>%
  addPolygons(data = bxl, color = 'grey')

#do a spatial join with the original bw_sf data frame to get the data back
bxl_voronoi_poly <- st_cast(st_buffer(bxl_voronoi, 0)) %>% 
  st_intersection(bxl) %>%
  st_sf() %>%
  st_join(bxl_measurements, join = st_contains)

Evaluation error: TopologyException: Input geom 0 is invalid: Self-intersection at or near point …

#create a palette (many ways to do this step)
colors <- colorNumeric(
  palette = 'Reds',
  reverse = FALSE,
  domain = bxl_voronoi_poly$value)

bxl_voronoi_poly <- bxl_voronoi_poly %>%
  mutate(popup_html = glue('<strong>Sensor ID</strong>: {sensor_id}<br/>
                           <strong>Location ID</strong>: {location_id}<br/>
                           <strong>Timestamp</strong>: {timestamp}<br/>
                           <strong>PM10-value</strong>: {value} µg/m3<br/>'))

#create the leaflet
m_bxl_vonoroi <- leaflet(bxl_voronoi_poly, width = '100%') %>% 
  #addProviderTiles('Stamen.Toner') %>%
    addTiles() %>%
  addPolygons(fillColor = colors(bxl_voronoi_poly$value),
              fillOpacity = 0.7, color = 'grey',
              weight = 1,
              popup = ~popup_html)

m_bxl_vonoroi